feat: Jamba tokenizer support (#86)

* feat: support get_tokenizer by tokenizer name

* fix: Added deprecation warning to count_tokens in ai21 client

* bump: ai21-tokenizer to 0.9.0

* test: Added tests for new tokenizer

* fix: model name in tests

* fix: test model name for chat

* fix: Added kwargs

* fix: type

* fix: doc

* docs: Count Tokens README

* fix: tokenizer example

---------

Co-authored-by: asafg <asafg@ai21.com>

Author: Josephasafg

pyproject.toml

@@ -57,7 +57,7 @@ packages = [
 [tool.poetry.dependencies]
 python = "^3.8"
 requests = "^2.31.0"
-ai21-tokenizer = "^0.3.9"
+ai21-tokenizer = "^0.9.0"
 boto3 = { version = "^1.28.82", optional = true }
 dataclasses-json = "^0.6.3"
 typing-extensions = "^4.9.0"